MAS-ProVe: Understanding the Process Verification of Multi-Agent Systems

TL;DR

This paper empirically investigates process verification methods for multi-agent systems built on large language models, revealing challenges in reliability, high variance, and the need for further research to develop more effective verification techniques.

Contribution

It provides a systematic empirical evaluation of various process verification paradigms and strategies for MAS, highlighting their limitations and performance gaps.

Findings

Process verification does not consistently improve MAS performance.

LLM-as-a-Judge generally outperforms reward-based methods.

There is a trade-off between context length and verification performance.

Abstract

Multi-Agent Systems (MAS) built on Large Language Models (LLMs) often exhibit high variance in their reasoning trajectories. Process verification, which evaluates intermediate steps in trajectories, has shown promise in general reasoning settings, and has been suggested as a potential tool for guiding coordination of MAS; however, its actual effectiveness in MAS remains unclear. To fill this gap, we present MAS-ProVe, a systematic empirical study of process verification for multi-agent systems (MAS). Our study spans three verification paradigms (LLM-as-a-Judge, reward models, and process reward models), evaluated across two levels of verification granularity (agent-level and iteration-level). We further examine five representative verifiers and four context management strategies, and conduct experiments over six diverse MAS frameworks on multiple reasoning benchmarks. We find that process-level verification does not consistently improve performance and frequently exhibits high variance, highlighting the difficulty of reliably evaluating partial multi-agent trajectories. Among the methods studied, LLM-as-a-Judge generally outperforms reward-based approaches, with trained judges surpassing general-purpose LLMs. We further observe a small performance gap between LLMs acting as judges and as single agents, and identify a context-length-performance trade-off in verification. Overall, our results suggest that effective and robust process verification for MAS remains an open challenge, requiring further advances beyond current paradigms. Code is available at https://github.com/Wang-ML-Lab/MAS-ProVe.